This project is designed to evaluate the role of DNA recombination in intrinsic radiation resistance. The cell killing effect of ionizing radiation is largely dependent upon the processing of double-stranded DNA damage and is an intrinsic property of the cell rather than an induced phenomenon. The genetic basis to radiation resistance is still poorly understood: ionizing radiation repair genes have been isolated, but their function has not been elucidated. The role of recombination in determining the response to ionizing radiation has been known in yeast, but the association in mammalian cells has only recently been realized. Scid mice, a phenotype caused by the lack of recombination within the immunoglobulin and T-cell receptor genes, have a general radiosensitivity of their cells including fibroblasts, and their cells lack the ability to rejoin DNA fragments caused by ionizing radiation. The current hypotheses are that recombination mechanisms have a major role in determining radiation resistance; and that tumorigenesis, which leads to an inherent instability of the genome and frequently leads to radiation resistance, has this effect mediated by a change in recombination function. Recombination will be evaluated by the use of integrating DNA vectors. The vectors will be designed to integrate into the genome, reported by the neo gene. Partial gpt sequences will constitute the test gene, which will require a recombination event to reconstitute the intact and functional gene (resistance to XHATM). To evaluate intra-molecular recombination events, the partial sequences will be separated along a single vector; for inter-molecular events, the partial sequences will be on separate vectors with a first round of transfection to establish an integrated target and a second round of transfection to complete the assay. Recombination events will be indicated by the loss of function of a Herpes Simplex Virus thymidine kinase gene, which will be located within the gpt sequences. This loss leads to resistance to ganciclovir. The frequency of recombination events will be indicated by the number of ganciclovir-resistant colonies; the proportion of accurate recombination events will be given by the proportion of XHATM-resistant colonies. Factors which could influence recombination parameters, such as double-strand breaks or gaps within the target, or specific sequences which appear to promote or mediate recombination will also be evaluated. The main focus of these assays will be human tumor cells, although characterization of the assays on well-studied repair deficient mutants is planned. Human tumors will be studied where differences in radioresistance exist, such as sensitive mutants, clonal heterogeneity, plus models of tumorigenesis and transfection of deleted tumor suppressor genes. A spectrum of clinical radiosensitivity will also be assessed.